Comparing field to remotely-sensed estimates of annual grass cover in Indaziflam treated areas

Authors: BARBER, COLE - University Of Idaho; Harrison, Georgia; STRAND, EVA - University Of Idaho

Submitted to: Society of Range Management
Publication Type: Abstract Only
Publication Acceptance Date: 1/1/2024
Publication Date: 2/1/2024
Citation: Barber, C., Harrison, G.R., Strand, E.K. 2024. Comparing field to remotely-sensed estimates of annual grass cover in Indaziflam treated areas. Society of Range Management. Abstract.

Interpretive Summary: The spread of annual grasses like cheatgrass has brought The spread of annual grasses like cheatgrass has brought significant changes to western rangelands. Understanding and accurately mapping the extent of annual grass cover is crucial to grasp the full scope of this invasion and its potential consequences, as well as to develop effective control strategies. One valuable tool for assessing plant cover across diverse landscapes is the Rangeland Analysis Platform (RAP). RAP combines satellite imagery with on-the-ground observations to create maps of plant cover and production. This project aimed to compare estimates of annual grass cover obtained through traditional field measurements with those generated by RAP in the same locations. Preliminary findings indicate that RAP's measurements of annual grass cover were within 10% of what was measured in the field. However, to fully understand the reliability of RAP as a tool for estimating annual grass cover, further investigation across various sites and years will be conducted.

Technical Abstract: Invasive annual grasses, such as cheatgrass (Bromus tectorum) have spread across western U.S. rangelands at a yearly rate of 14%. Annual grass invasion results in altered fire regimes, loss of forage, and habitat for wildlife and livestock. Recently, a new herbicide, Indaziflam (Rejuvra), has shown promising results for controlling invasive annual grasses. The ability to monitor annual grass cover, including in areas receiving Indaziflam treatment is crucial. Traditional field methods for estimating plant canopy cover are well suited for collecting highly detailed data at fine spatial resolutions but can be limited in their ability to extrapolate to larger scales in heterogenous landscapes. Remote sensing tools such as the Rangeland Analysis Platform (RAP) provide estimates of cover and production by plant functional group and can fill in spatial and temporal gaps of traditional field monitoring. While these tools have high utility for examining historical trends in plant cover across variable landscapes, there are still questions of t how RAP compares to field estimates. In this research, we compare RAP and field-estimated annual grass cover across three spatial scales. Field-estimated data are from independent monitoring in Idaho and Wyoming of areas aerially treated with Indaziflam from 2011-2021. Preliminary results suggest that RAP estimates of annual forb and grass cover are within a range of 10% of field measurements, which is higher than mean absolute error (7%) reported by RAP for this functional group.  Differences between field and RAP estimates were greater at larger scales. These results have implications for the use of remotely sensed estimates of fractional cover for rangeland vegetation monitoring.